PhD position on learning with algorithmic technologies in medical diagnosis

at  Vrije Universiteit Amsterdam

YOUR FUNCTION

Vrije Universiteit Amsterdam is offering a fully funded, four-year PhD-position on learning with algorithmic technologies, as part of the NWO-funded project “Learning in artificial intelligence across and beyond occupational communities (LIAISON): Towards effective deployment of novel algorithmic technologies in medical diagnosis”.
Sophisticated algorithms, machine learning, deep learning, and Artificial Intelligence (AI) enable innovative ways of working and learning. A concrete and promising example of this is the use of AI in the domain of radiology. By using AI for diagnosing, radiologists can reduce their day-to-day workload, freeing up time to spend on medically challenging cases. This in turn can lead to more rigorous diagnostic accuracy, increased senses of autonomy in managing one’s daily work.
However, the current use of algorithmic technologies for making such improvements in medical practice has been limited, short-sighted, fragmented, and unsystematic. While organizations and institutions are investing significantly in algorithmic technologies, this does not yet create sustainable, significant, viable values for patients, society, and the healthcare ecosystem.
Learning communities can help to overcome many problems by providing a safe and effective space for various stakeholders to collaboratively define, experiment with, and reflect on learning scenarios. In this project, we envision a learning community around ways of using algorithmic technologies in medical diagnosis, where developers, medical professionals, patients, and policy makers come together to:
1) explore and identify a range of possibilities for using algorithmic technologies for medical diagnosing, thereby considering different perspectives including technical, social, and ethical perspectives.
2) collaboratively experiment with algorithmic technologies in use-cases, thereby training workers in the efficient and effective development, implementation, and use of such technologies for medical diagnosing.
3) continuously share experience and insights to improve algorithmic technologies for medical diagnosing.
4) translate experiences from use cases into concrete learnings to provide input for education and training programs.
As PhD candidate, you will focus on understanding how people learn with algorithmic technologies within and across specific use-cases. You will be responsible for collectively designing and implementing experiments around use cases of medical diagnosing.
You will be part of a team of two PhD candidates and three senior researchers (Dr. Mohammad Rezazade Mehrizi; Dr. Wendy Günther and Dr. Willem Grootjans) at Vrije Universiteit Amsterdam and Leiden University Medical Center, and work with a community of medical professionals, patients, developers, and policy makers.

You will not be expected to train algorithms or engage in data science work. Rather, you will focus on understanding how people learn with algorithmic technologies.
You will need to develop a thorough understanding of learning theories and computer-human interaction. You will need to apply knowledge of how people experience, engage, and disengage in different types of learning traps. You will design and implement experiments around specific use-cases of using algorithmic technologies for medical diagnosing. In doing so you will set up and design systems, platforms, and interfaces required for running the experiments. In doing so, you will be considering the wishes of various stakeholders, including medical professionals, patients, developers, and policy makers. Through the experiments, you will capture and analyze quantitative and qualitative data on different types of learning occurring as humans interact with algorithmic technologies. You will also be expected to actively share insights and transfer research outcomes.
PhD candidates will give a modest contribution to the department’s teaching program, with the opportunity to obtain a basic qualification teaching certificate (shortened BKO).